Publication detail

Dynamic Soaring in Uncertain Wind Conditions: Polynomial Chaos Expansion Approach

NOVÁK, J. CHUDÝ, P.

Original Title

Dynamic Soaring in Uncertain Wind Conditions: Polynomial Chaos Expansion Approach

Type

conference paper

Language

English

Original Abstract

Dynamic soaring refers to a flight technique used primarily by large seabirds to extract energy from the wind shear layers formed above ocean surface. A small Unmanned Aerial Vehicle (UAV) capable of efficient dynamic soaring maneuvers can enable long endurance missions in context of patrol or increased flight range. To realize autonomous energy-saving patterns by a UAV, a real-time trajectory generation for a dynamic soaring maneuver accounting for varying external conditions has to be performed. The design of the flight trajectory is formulated as an Optimal Control Problem (OCP) and solved within direct collocation based optimization. A surrogate model of the optimal traveling cycle capturing wind profile uncertainties is constructed using Polynomial Chaos Expansion (PCE). The unknown wind profile parameters are estimated from observed trajectory by means of a Genetic Algorithm (GA). The PCE surrogate model is subsequently utilized to update the optimal trajectory using the estimated wind profile parameters.

Keywords

Polynomial Chaos Expansion, Surrogate Modeling,  Dynamic Soaring, Optimal Control

Authors

NOVÁK, J.; CHUDÝ, P.

Released

16. 2. 2024

Publisher

Springer Nature Switzerland AG

Location

Grasmere

ISBN

978-3-031-53968-8

Book

Machine Learning, Optimization, and Data Science

Edition

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ISBN

0302-9743

Periodical

Lecture Notes in Computer Science

Number

14505

State

Federal Republic of Germany

Pages from

104

Pages to

115

Pages count

11

BibTex

@inproceedings{BUT185184,
  author="Jiří {Novák} and Peter {Chudý}",
  title="Dynamic Soaring in Uncertain Wind Conditions: Polynomial Chaos Expansion Approach",
  booktitle="Machine Learning, Optimization, and Data Science",
  year="2024",
  series="Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
  journal="Lecture Notes in Computer Science",
  number="14505",
  pages="104--115",
  publisher="Springer Nature Switzerland AG",
  address="Grasmere",
  doi="10.1007/978-3-031-53969-5\{_}9",
  isbn="978-3-031-53968-8",
  issn="0302-9743"
}